Ghost-Busting Heat Transfer In Return Air Plenums
Common sense, a bit of heat transfer knowledge and strict adherence to specifications go a long way to improving the air/hydronic dynamic.
One of the landmines associated with radiant based HVAC systems using fan/coils for humidification and decontamination is the disconnect often created between those responsible for the air system and those responsible for the hydronic system. Both of them are installing heat exchangers of one form or another. These heat exchangers are intimately connected by the on-site assembly of their respective “stuff.”
One hazard that has crept up occasionally over the years has to do with the poor practice of flashing joist bays for return air plenums. When air is circulated through an un-ducted and un-insulated return air plenum under a heated or cooled floor, the laws of heat transfer will definately tell you that you are about to have an expensive problem to solve.
Here is what can happen: the call comes in from the client complaining about discomfort, so you head on over to the job site. Lo and behold your client has a case. You start your diagnostics and discover the supply air being delivered to the space is 15F (8.3C) higher than design setpoint of 72F (22C). You discover that a control valve you expected to be stuck open is in the closed position.
Now you ask yourself where is the air picking up heat if the valve is closed. One possible source is valve leakage. I know it is a shock but valves leak, some more than others. Even with a valve in the closed position and its actuator torque of sufficient strength to close off against pump head – yes you can still have leakage.
How much? Well, it all depends, but in this case the leakage was too small to be of any significance so that was not the prime problem. You think to yourself it has to be coming from the radiant floor, but how could that be when your specs called for insulated ducts? Your brain is now doing mental gymnastics until the light comes on.
“I bet the sheet metal guy didn’t look at the spec for insulated ductwork and just flashed the bottom of the joists,” cause that is what a seasoned ghostbuster would think. You have the general contractor take down a small piece of drywall and your suspicions are confirmed (see Figure 1).
Here is what is going on. The radiant system is connected to the air plenum first by heat being conducted to the underside of the sub floor. Wood or fake wood products have rough surfaces and high emissivities.
If you want to scrub away heat from a heated floor with air, plus extract energy in the form of heat with radiant, nothing serves as a better surface than a subfloor. Additionally, there could be any combination of air flow characteristics depending on the velocity and that will have an impact on the convective heat transfer.
You also have a lower emissivity/higher reflectivity sheet metal on the bottom so the radiant energy emitted to the flashing and sides of the joist from the heated sub-floor is going to be bounced back up to the sub-floor where it will be absorbed and reradiated. This further contributes to the increasing temperature in the joist cavity. If you think about it – what you have discovered is a radiant ceiling system with forced draft, as such the return air has no choice but to increase in temperature even if the control valve on the fan/coil is closed.
An overview of the heat transfer mechanisms at play for un-insulated and insulated return air plenums is shown in Figure 2. It can be expressed in standard heat transfer notation where U is the combined radiant and forced convection heat transfer coefficient with the log mean temperature differential determined by Figures 3, 4 and 5.
Since the amount of heat transferred is connected to the surface area (A), the X dimension in Figure 2 becomes really important; that is, the longer the X value the greater the heat transfer (q) value in Figure 5.
The end conversation for this ghostbusting always goes like this: Mr. Sheet Metal says to owner, “Well it’s obvious, turn off the radiant floor.” The angry client says, “I paid big bucks for my radiant floors so you know what you can do with your suggestion.” You step in between the forces of good and evil and explain to the general contractor and the sheet metal contractor that you need the air for humidification and decontamination and the radiant for thermal comfort. You are going to have to pull down the ceiling and have Mr. Sheet Metal rip down the flashing and install as per specs with the insulated ducting.
Ghost heat transfer is not a trivial thing in any system, but it is particularly worrisome with hybrids. Not only does it contribute to instability in thermal comfort, higher return air temperatures destroy coil heat exchanger efficiencies. It has the same effect as installing an oversized valve, which leads to a destruction in valve authority, resulting in valve hunting and coil over performance during low loads.
Although there is no hard and fast rule of thumb to use, my current standard specification calls for one layer of R5 under the floor, plus fully wrapped to R5 external duct insulation on shorter return air plenums under heated floors constructed with masonry surfaces. For longer runs under floors with less conductive flooring, I will bump up the first layer to R10. I will also take into consideration design fluid temperatures, tube density and patterns over the return air plenum.
No one said this hybrid stuff was easy but with some common sense, a bit of heat transfer knowledge and strict adherence to your specifications, even the air guy and hydronic guy can get along. <>
engineering through the Association of Professional Engineers, Geologists and Geophysicists of Alberta.
Print this page
May 11, 2022